CA2149767A1

CA2149767A1 - Membranes having improved selectivity and recovery, and process for making same

Info

Publication number: CA2149767A1
Application number: CA002149767A
Authority: CA
Inventors: John A. Jensvold; Srikanth R. Chary; Wendy S. Jacks; Hans R. Keller; Theodore L. Parker; Damoder Reddy
Original assignee: Individual
Current assignee: Dow Chemical Co
Priority date: 1992-12-01
Filing date: 1993-11-19
Publication date: 1994-06-09
Also published as: EP0671971A1; DE69319133D1; JPH08503655A; ATE167081T1; EP0671971B1; DE69319133T2; WO1994012269A1

Abstract

The invention is a process for obtaining membranes having improved selectivity and recovery using a combination of heat treating and UV irradiation.

Description

~s WO 94/12269 PCT/US93/11;!75 MEMBRANES HAVING IMPROVED S~LECTIVITY AND RECOVERY, AND PROCESS FOR MAKING SAME

The invention relates to gas separation membranes having improved selectivity 5 and recovery, and a process for making same.
Gas separation membranes have been used for some time to separate a number of gases from gas mixtures. A tradeoff typically exists between selectivity and permeabi I ity.
Membranes which possess high permeabilities generally have low selectivities, while membranes which possess high selectivities generally have low permeabilities.
10For example, in the separation of oxygen from nitrogen, while some membranes have been reported in the literature to have oxygen/nitrogen selectivities of 3 to 8, the corresponding permeabilities are s: nly in the range of from 100 to 0.1 barrers. Thus, the recovery, that is, the ratio of the volume of non-permeate product gas to the volume of feed gas, of such mernbranes is low, which limits the economic usefulness of such membranes, l ~i particularly when a high purity non-permeate product such as 99 percent nitrogen is desired.
What is needed is a membrane which possesses both high selectivity and moderate to high permeability. Particularly as higher purity non-permeate product gas is required, membranes possessing high ecovery are needed. If permeability can be maintained, membranes with higher selectivities are capable of achieving a higher recovery at a given non-2~ permeate concentration or purity than membranes with lower selectivities. What is therefore needed is a process for increasing the selectivities of membranes which have moderate selectivities and moderate permeabilities, without a corresponding significant decrease in permeabilities, so that high recoveries can be obta.ned.
The invention is a process for treating a gas separation membrane to improve its25 selectivity, comprising:
A. first heating a membrane comprising a polymer having a UV excitable site and a labile protonic site in the polymeric backbone, such that a covalent bond may be formed between said sites, at a temperature between 60 and 300C for a time sufficient to relax excess free volume in the polymer; and B. then irradiating the membrane with a UV radiation source in the presence of oxygen for a!~ime sufficient to surface oxidize the membrane to obtain a treated membrane;
wherein the treated membrane exhibits at least a 10 percent increase in selectivity with less than a 60 percent decrease in permeability compared to the untreated membrane.
The invention includes membranes prepared from the above-described process.
Such membranes possess improved selectivity and recovery compared to untreated membranes.

WO 941~69 , ~ PCT/U593/11275 Membranes useful in this invéntion include those derived from polymers containing a UV excitable site and a labile protonic site in the polymeric backbone such that a covalent bond may be formed between said sites.
While not wishing to be bound by any particular theory, it is believed that the 5 improvement obtained in selectivity after exposure to heat and Uv irradiation is related to the presence of hydrogen atom donors and to the ease of hydrogen atom abstraction by a UV
excitable site to farm free radical species in the presenc~ of molecular oxygen. Generally, the more readily an hydrogen atom is removed, the better the efficacy of the heat and UV
irradiation treatment should be and the concomitant improvement in selectivity expected. In a 10 thermodynamic sense, the lower the bond dissociation energy, the more read;ly the hydrogen atom may be abstracted. In general, benzylic and allylic hydrogens are more easily removed, followed by tertiary, then secondary, and fi nally primary and aromatic hydrogens.
Additionally, hydrogen atoms bonded to carbon atoms that are alpha to a carbonyl group or an ether group are also readily abstracted. It is also believed that some degree of interchain bonding or cross-linking may occur by reaction of the free radical species generated. In determin;ng which types of polymer structures would be expected to give favorable improvements in membrane selectivity after exposure to heat and UV irradiation, two factors must be considered: (l) the type and number of hydrogen atom donor functions present, and (2) the type and number of groups or structures capable of absorbing the UV radiation and 20 efficiently producing free radical species.
Membranes useful in the process of this invention include those which are derived from poiymers containing hydrogen atom donor sites and structures capable of absorbing the UV irradiation including carbon/hetero atom double bonds~uch as keto, ester, carbonate, amide, sulfoxyl, sulfonyl, and nitro moieties, aromatic/conjugated double bonds such as C = N6 25 and pairs of highly polarizing groups (elec~ron donatinglaccepting) such as -R, -OR, and -SR, and -CN, nitro, sulfonic, carboxylic, ester, and halogen.
Preferably, the membrane comprises an unsubstituted or s~lbstituted polycarbonate, polyestercarbonate, polyester, polystyrene, polysulfone, polyethersulfone, polyether, polyarylester (polyarylate), polyethylene terephthalate, cellulose ester, 30 polybenzazole, polyurethane, or copolymer o- physical blend thereof. More preferably, the membrane comprises an unsubstituted or substituted polycarbonate, polybenzazole,polyestercarbonate, polyester, polysulfone, polyethersulfone, polyether, polyarylester (polyarylate), or copolymer or physical blend thereof~ Even more preferably, the membrane comprises a polymer having in general a rigid structure such as a polybenzazole or a polymer 35 containing 9,9-bis(3,5-dibromo-4-hydroxyphenyl)fluorene, 9,9-bis(3,5-dichloro-4-hydroxyphenyl)fluorene, hexafluorobisphenol A, tetrahalohexafluorobisphenol A, or tetraalkyl-hexafluorobisphenol A moieties selected from the group consisting of polyether, wo g4,~69 2 1 ~ 9 7 6 7 ~ PCr/U593/11~75 1~

polysulfone, polyethersulfone, polyarylester (polyarylate), polyester, polyestercarbonate, polycarbonate, and copolymers and physical blends thereof.
Polybenzazole (P8Z) as used herein means a polybenzazole polymer selected from the group of polybenzoxazoles and polybenzobisoxazoles (PBO), polybenzothiazoles and 5 polybenzobisthiazoles (PBT~, and polybenzimida~oles or polyben~obisimida~oles (PBI) The term polybenzoxazole (PBO) refers broadly to polymers in which each mer unit contains an oxazole ring bonded to an aromatic group, which need not necessarily be a benzene ring The term polybenzoxazole (P80) also refers broadly to poly(phenylene-ben?obisoxazole)s and other polymers wherein each mer unit comprises a plurality of oxazole rings fused to an 10 aromatic group. Similar meanings shall apply to the terms polybenzothiazole (PBT) and polybenzimidazole (PBI). Hexafluorobisphenol A as used herein means bisphenol A wherein all six of the hydrogens on the isopropylidene bridging group have been substituted with fluorine moieties. Tetrahalo or tetraalkyl as used herein with respect to bisphenol A means bisphenol A
- wherein four of the eight hydrogens on the aromatic rings are substituted with halo or alkyl moieties respectively.
The membranes may have morphological structures which are non-porous, asymmetric (anisotropic~, or composite. Non-porous membrane as used herein means a membrane whish is dense, that is, substantial Iy free of holes or voids. Asymmetric membrane as used herein means a membrane which possesses at least one discriminating region and at 20 least one porous region, wherein the discriminating and porous regions comprise the same polymer. Composite membrane as used herein means a membrane which possesses at least one discriminating layer and at least one porous layer, wherein the discriminating and porous layers comprise different polymers.
The membranes may be in the configuration of hollow fibers or tubes, or films or25 flat sheets~
The membranes are first treated by heating the membrane to retax excess free volume in the polymer. The heating temperature is preferably between 60 and 300C, more preferably between 100 and 250C, even more preferably between 150 and 230C. The optimum heating temperature for a given polymer is in the range of Tg - 200C to Tg - 50C, 30 wherein Tg is the glass transition temperature of the polymer. The heating time is preferably between O.S and 24 hours, more' pre~erably between 0.5 and l 2 hours, even more preferably between 0.5 and 4 hours.
The membranes are then irradiated with UV radiation in the presence of oxygen such that the surface is at least partially oxidi ed. A UV source having radiation wavelengths 35 between 180 and 400 nanometers is preferred, between 200 and 375 nanometers is more - preferred. The exposure time is preferably between 1 and 90 minutes, more preferably between 3 and 80 minutes, even more preferably between S and 60 minutes. In general, WO 94/12269 PCTIUS93/11275 ~;3 21~9767 ~

preferred UV irradiation times for composite and asymmetric membranes are significantly shorter than for homogeneous (dense) membranes for a given polymer The treated membranes preferably possess an increase in selectivity of at least about 10 percent compared to the un~reated membrane for at least one gas pair selected from 5 the group ~onsisting of oxygen/nitrogen, carbon dioxide/methane, hydragen/light hydrocarbon, heliumlmethane, and nitrogen/methane. The treated membranes more preferably possess an increase in selectivity of at least about 30 percent compared to the untreated membrane for at least one gas pair selected from the group consisting of oxygeninitrogen, carbon dioxide/methane, hydrogen/light hydrocarbon, helium/methane, and 10 nitrogen~methane. The treated membranes preferably possess a selectivity for oxygen/nitrogen at 30C of at least about 6, more preferably of at least 7.5, even more preferably at least 9.
The treated membranes preferably exhibit a decrease in permeability of less than- about 60 percent compared to the un~reated membrane, more preferably of less than about 40 j 5 percent compared to the untreated membrane, even more preferably of less than about 25 percent compared to the untreated membrane, for at least one gas selected from the group consisting of helium, oxygen, nitrogen, carbon dioxide, methane, hydrogen, and a light hydrocarbon. The treated membranes preferably possess an oxygen permeability of at least 0.3 barrers, more preferably of at lea~t 0.5 barrers, even more preferably of at least 1.0 barrers.
;~0 The treated membranes may be fabricated into hollow fiber, tubular, spiral wound, or plate and frame devices by methods known in the art.
The membranes of this invention are useful for separating one or more gases from gas mixtures, including mixtures comprising hydrogen, helium, oxygen, nitrogen, air, argon, carbon monoxide, carbon dioxide, ammonia, water vapor, light hydrocarbons, natural 25 gas, hydrogen sulfide, nitrogen oxides, sulfur oxides, andlor organic vapors. Light hydrocarbons as used herein means saturated and unsaturated C1-4 hydrocarbons~
The membranes of this invention are useful for many different gas separation applications, ;ncluding, but not limited to, providing of an enriched nitrogen stream for inerting of flammable fluids, perishable foodstuffs, and heat treatment of metals, providing an 30 enriched oxygen stream for medical uses, fermentation processes, or enhanced combustion processes, recovering carbon dioxide from light hydrocarbons, treating flue gases to remove nitrogen oxides and sulfur oxides, removing organic vapors from air, dehydrating air and natural gas~
For gas separation, the ~perating temperature for such applications is preferably 35 between -20 and 100C, more preferably between S and 60C, and the operating pressure is preferably between l O and 1000 psi. (69 and 6895 kPa), more preferably between 50 and 500 psi. (345 and 3447 kPa).

~;~ WO 94/122~9 2 1 4 9 7 6 7 P(~T/US93111~7 The membranes of this invention may be operated in any configuration or combination, including, but not limited to, parallel, series, recycle, or cascade operations.
These hollow fiber membrane devices may also be operated in conjunction with other separation processes or unit operations including, but not limited to, cryogenic distillation, 5 pressure or temperature swing adsorption, liquid absorption The following examples are for purposes of illustration only and are not intended to limit the scope of the invention or claims.
Examp!e 1 Treatment of Tetrabromobisphenol A Polvcarbonate Membranes At StandardConditions Casting solutions were prepared from mixtures containing about 15.00 grams tetrabromobis,ohenol A polycarb-~nate (TBBA PC) in about 75.00 grams methylene chloride and about 10.00 grarns 1,2-dichloroethane. The mixtures were stirred for 2-12 hours to form substantiaily uniform solutions which were filtered through 3 micron filters. The polymer solutions were cast onto clean glass plates with a 6 or 10 mil doctor blade. After casting, the 15 films were covered within about 5 seconds with a second clean glass plate, using metal washers as spacers between the plates. The entire casting assemblies were then covered with an air diffuser box to cause slow, even evaporation of solvent from the cast fiims. After about 2 hours at ambient temperature, the glass plates with the films were immersed in ambient temperature distilled water to facilitate film separation from the plates. The films were then dried in an 20 oven for about 2 hours at about 50C.
The dried films, which had thicknesses of 10 to 60 microns, were used to make sample discs about 1.7 inches (4.4 centimeters) in diameter. Samples were reserved as controls, whi le other samples were subjected to heat treatment andror UV irradiation treatment.
Heat treatment conditions were for about 2 hours at about 180C and UV
25 irradiation exposures were for about 20 minutes. The UV radiation source having a wavelength of 200-600 nanometers was an ACE l~anovia mercury vapor lamp with reflector having a 450 Watt power supply. The membrane samples were placed at a distance of about 22 centimeters from the UV radiation source and irradiated on one side only. The radiant power striking the samples was about 100 mW.
The membranes were then eval uated for the separation of oxygen/nitrogen usi ng the constant-volume variable-pressure method. The feed side of the membrane was pressurized with a single gas (oxygen or nitrogen) at about S0 psig and at 25C as the feed gas Data are reported for the samples in Table 1. The membranes which received both heat treatment and UV radiation treatment possessed signi ficantly higher selectivities than the 35 control membrane, or the membranff which received only heat treatment or only UV
irradiation treatment. In addition, the heat and UV treated membranes exhibited minimal loss in permeab;lity compared to the control membrane.

W094/~6g 214 9 7 6 7 PCT~S~3/1127S ~

` TABLE I
Tetrabromobisp'henol A Polycarbonate Membranes , _ . , _ _ _ . . _ Sample Treatment Oxygen/Nitrogen (Barrers) . . . _ . ............ _ 1A Control 6.90 1.08 . _ _ . .
1B Heat at 7.40 0.94 180C only , , ., ..... ~__ ~
1C UV at 20 7~94 1.00 min. only . . . .. .
1D Heat at 9.26 0.88 180C + UV
at 20 min.
. . . _ ~ . .
1AA Control 6.97 1 1~' . . . _ _ _ .,, _ 1BB Heat at 8.18 0.99 180C only 1 5 . . , .. . . . - . . , . , _ 1CC UV at 20 7.87 1.01 minO only . . . . . . .. , 1DD Heat at 9.26 0.90 180C + UV
at 20 min.
. . . . - _ . .

Exam~le 2-Treatment of Tetrachlorohexafluorobis~henol A Polvcarbonate Membranes At Standard Conditions Tetrachlorohexafluorobisphenol A polycarbonate (TCHF PC) membranes were 2~; prepared and treated in a manner similar to that described in Example 1. Data are reported for - the samples in Table 11. The membranes which received both heat treatment and UV irradiation treatment possessed significantly higher selectivities than the control membrane, or the membranes which received only heat treatment or only UV irradiation treatment. In addition, the heat and UV treated membranes exhibited only a 24 percent decrease in permeability ~ WO94/L~69 PCT~S93/11275 2149767- ~
compared to the control membrane.

TABLE II
Tetrachlorohexafluorobisphenol A Polycarbonate Membranes . ~ ~

Sampie Treatment Oxygen/N1trogen Permeability 2A Control ~ 4.95 8.99 2B Heat at 5O72 7.86 180C only .... .. ... . , . _ 2C UV at 20 5.22 9.65 min. only . - . , ., . . . . - . 1 2D Heat at 5.96 6.80 180C + UV
.. _ at 20 min~ .

Example 3 ~Treatment of Polvethersulfone Membranes Membranes from polyethersulfone, Victrex 600P from ICI, were prepared and treated in a manner similar to that described in Example 1, except that the UV irradiation time was 4S minutes. Data are reported for the samples in Table lli. The combined heat and UV
irradiation treatment increased the selectivity significantly over the untreated membrane, with only a slight decrease in permeability.
TABLE III
Polyethersulfone Membranes Sample Treatment Selecti~ity Per--ab~lity . - . . . .~
3A Control 5.79 0.47 3B Heat at 3.06 0.53 _ _ ~ 780C only . . ~
3C U~ at 45 ___ ___ ~min. only ~ __ _ _ Heat at 9.38 0.36 35 at 45 min. . ~

W094112269 21~9767 PC~rtJS93/1127~ ~ , Example 4--Treatment Of Polvsulfone Membranes Membranes from polysulfone, Udel P1700 fronl Amoco, were prepared and treated in a manner si milar to that described i n Example l, except that the heat treatment temperature was either 120C or 160C, and the UV irradiation time was 45 minutes. In 5 addition, for Sample 4E, the order of treatment was reversed: the membrane was first UV
irradiated for 30 minutes, and then heat treated at 1 20C. Data are reported for the samples in Table IV. The combined heat and UV irradi~tion treatment resulted in membranes with higher selectivities than heat treatment alone. The membrane which was UV irradiated and then heat treated (4E~, did not exhibit the high selectivity of the sample which was first heat treated and 10 then UV irradiated ~4G).

TABLE IV
Polysulfone Membranes _ _ __ ., Sample Treatment Selectivity (Barrers) , .
4A Control __ __ _ , , _, , _ _ _ 4B Heat at 6.27 0.62 120C only 4C UV at 30 7.47 0~61 : min. only 4D Heat at 7.87 0.49 120C ~ UV
at 45 min.
. _ .
4E UV at 30 5.98 0.57 min. + heat at 1~0C
.
4F Heat at 5.43 0.76 160C only _ . .
4G Heat at 7.80 0.51 160C + UV
at 45 min.

.;
: !

t~G~ WO 94/1~269 PCT~S93111275 ~, ,,.. , -Example 5--Treatment Of Pol~stvrene Membranes Membranes from polystyrene, Styron 685 from Dow Chemical, were prepared and treated in a manner similar to that described in Example l, except that the heat treatment temperature was 80C and the UV irradiation time was either 30 minutes or 60 minutes. Data 5 are reported for the samples in Table V. The combined heat and UV irradiation treatment resulte~l in membranes with higher selectivities than heattreatment alone.

TABLE V
Polystyrene Membranes . . . ......... _ --L Sample ¦ 1l ea::men~
Selectivity ~ r 5A Con tro l 5 . 9 4 2 . 4 9 5B Heat ~t 6 . 30 1 ~ 79 80C only _ , _ _ _ 5C UV at 30 6 . 26 1 . 98 . . mln. only . _ .
5D Heat at ~ . 51 1 . 72 80C + UV
. . . at 30 min. . -5E Heat at 9 . 30 1 . 17 80C + UV
. - . at 60 min.

25 Exam~le 6--Treatrnent Of Polvester Mernbranes Membran~s from polyester, Ardel D-100 from Union Carbide, were prepared and treated in a manner similar to that described in Example 1, except that the heat treatment temperature was 1 OO~C and the UV irradiation time was 45 minutes. Data are reported for the samples in Table Vl. The combined heat and UV irradiation treatment resulted in membranes 30 with higher selectivities than heat treatment alone.

g WO ~4/12269 PCT/US93111275 214g76~ , `
TABLE V I
Polyester Membranes _ -.... ,.. _ S Sa~ p 1 e Tr ea tmen tSe 1 ec t i v i t y1~e r m e a ~ y 6A Control __ . _ .
6B Heat at 4 . 62 1 . 04 . . ~ 1 0 0C on 1 y ~ . .
6C Heat at 4 . 86 O . 92 100C + UV
. . at 45 min. _ _ . . _ Example 7--Treatment Of Bis~henol AP Polvcarbonate Membranes - Bisphenol AP polycarbonate (BAP PC) may be synthesized bythe following 15 procedure. To the reactor was added water, about 300 milliliters, Bisphenol-AP, about 5.0 grams, para-tert-butylphenol (PTBP), about 0.827 grams, methylene chloride, about 188 milliliters, and 50 percent NaOH, about 33.8 grams. Reaction pH was 13.5 and was maintained at 12.2 - 12.7 throughout the phosgenation by adding 50 percent NaOH as necessary. The reaction rnixture was vigorously stirred and phosgene, about 26 grams, was added. After the 20 phosgene had been added, a drop of the aqueous phase was placed on a petri dish and acidified to a pH of 1 with 1 N HCI. A precipitate indicates that monomer was still present, so more phosgene was added until there was no precipitate. ~
Next, a drop of the organic phase was combined with one drop of 0.25 percent nitrobenzylpyridine in tetrahydrofuran. Orange color formation indicated the presence of 25 chloroformates, whereas a colorless solution indicated no chloroformates. Assuming chloroformates were present, additional methylene chloride, abo~t 188 milliliters, was added - to the reaction mixture followed by triethylamine, about 120 milliliters. The reaction mixture was allowed to stir for 10 minutes while maintaining the pH at about 12Ø The organic layer was analyzed for chloroformates and if there were none, the reaction pH was reduced to 2, by 30 carefully adding 1 N HCI. (At a pH of about 8-9, bicarbonate begins decomposing to give off C02, so caution is advised.) The aqueous phase was separated off and the organic iayer washed with water, 2 volumes of about 250 milliliters. The washed polymer was recovered using a hot water devolatilizer. Membranes from BAP PC were prepared and treated in a manner similar to that described in Example l, except that the membranes were UV treated only. Data are 35 reported for the samples in Table Vll.

~ wos4lL~69 PCT~S93/1~75 TABLE VII
Bisphenol AP Polycarbonate Membranes ~ , . _ . . ~
Sample Treatment Oxygen/Nitrogen (Barrers) . . ...... . - ......................... .-._ 7A Control 5.63 1.29 .. . . . . , . _. _ . .
7B UV at 20 5.96 1.05 - ~- ...... min. only . . . -.

Exarn~le 8--Treatment Of Polybenzoxazole Membranes Polybenzoxa~ole (PB0) may be synthesized by the following procedure. A
mixture of about 258 grams of 81 percent polyphosphoric acid (PPA) and 10.00 grams of diaminoresorcinol dihydrochloride (DAR) was placed in a 500 milliliter resin kettle. The resin 15 kettle was equipped with a nitrogen inlet, silicone oil heating bath, stainless-steel stirring shaft, and a high-torque stirrer. The mixture was heated at about 11 0C for 16 hours. At that time, about 89 grams af phosphorous pentoxide ~P205) and 15.23 grams of 1,1,3-trimethyl--3-phenylindan~,5'-dicarboxylic acid ~PIDA) was added. The reactants were heated according to the following schedule: about 7 hours at 1 1 0C, 16 hours at 1 50C, 24 hours at 1 80C, and 2~ 24 hours at 1 90C. The crude polymer was isolated by preci pitating into water, vacuum filtrating, washing with hot water and methanol, and finally drying in a vacuum oven. The polymer was soluble in m-cresol, trifluoroacetic acid, and~nethane sulfonic acid. PB0 membranes were prepared and treated in a manner similar to that described in Example 1, - except that the membranes were UV irradiated for 10 minutes only. Data are reported for the 25 samples in Ta~le Vlll.
TABLE VIII
Polybenzoxazole Membranes . . . ..,.. _ , ..

¦ Sample ¦ Treat~3nt ¦x5geleen/Nl~irt en¦Permeability¦

8A Control 5.34 12.25 8B UV at 10 6.21 9.11 min. only W094l~2~9 PCT~S93/1~75 ~

~xample 9--Treatment of TB8A PC Men~branes At Various Conditions TBBA PC membranes were prepared and ~reated in a manner similar to that described in Example 1, except tha~ the rnembranes were heat treated at different temperatures and irradiated at different times. Data are reported in Table IX.
TABLE IX
Tetrabromobisphenol A Polycarbonate Membranes . ~ , _ . . ._.
Heat . Oxygen/ Oxygen Sample Temperature UV tlme Nitrogen Permeability C (minutes) Selectivity (Barrers) . ~ . __ , . __ 11A 25 O 6.931.16 _ __ ~ _ _ 11B 25 30 each 7.570.57 s ide . . . _ 11C 60 O 5.911.25 r . ___ _ _ .
11D 60 20 6.871.11 ~ _ . . .
., 11E 60 30 6.72 1.16 _ __ 1lF 120 O 6.78 1.13 . . , . 11G 120 20 7.44 0.97 11H 12Q 30 7.44 0.94 . - . . _ .
11I 180 O 7.98 1.17 11J _ 1 15 8.37 1.20 . - . _ _ 1lK 180 20 ~ 8.23 1.03 __~ _ 11L 180 30 B.75 1.06 __ . .
11M 180 35 9.06 0.75 . .. ,_ _ _ 11N 180 ~5 9.41 0.77 _ . ._ .
110 180 30 each 9.90 0.74 side , _, _ , .
11P 180 20 one 10.19 0.70 side + 30 each side . . i _ _~ _ 11Q 230 O 8.69 1.05 11R 230 9 11.52 0.59 t~-~ WO 94/12269 214 9 7 6 7 ~ PCT/US93/11275 Comparative Example 10--Treatment Of TB8A PC Membranes Bv First UV Irradiatinq And Then Heatinq This Example is for cornparative purposes and is not illustrative of the invention.
TBBA PC membranes were prepared in a manner similar to that described in 5 Example 1. However, the membranes were first exposed to UV i rradiation for 30-60 minutes and then heat treated for about 2 hours at about 1 80C. Data are reported in Table X. The membrane samples which were first exposed to UV irradiation and then heat treated did not differ significantly in selectivity or permeability compared to the control samples.

TABLE X
Tetrabromobisphenol A Polycarbonate Membranes ~", ~ _ , . .~
Sample (minutes) Heat Oxygen/ (Barrers) 1 5 , ~_ , . ~ , .
12A 0 25 7.98 1.17 12B 30 180 7.50 1.36 ~ r l_ _ ~
12C 40 180 8.13 1.06 . _ ....
12I) 45 180 7.94 1~ 13 _, 12E side 180 8.03 .

Com~arative ExamDle 11 Treatment Of TBBA PC Membran_ Bv UV l!radiatina Under Nitroaen This Example is for comparative purposes and is not illustrative of the invention.
TBBA PC membranes were prepared and treated in a manner similar to that desaibed in Example 1, except that UV irradiation was performed under a nitrogena~:mosphere. Data for the samples are reported in Table Xl. The membrane samples which ;~ were UV irradiated under nitrogen did not differ signifi~antly in selectivity or permeability 30 compared to the control samples.

' ' `

i !

~- -1 3-:j `~

W094/~69PCT~S93/11275 ~
- 21~9767 TABLE XI
Tetrabromobisph~nol A Polycarbonate _ Me nbranes Oxygen/ Oxygen 5Sample Treatment Nitrogen Permeability Selectivity (Barrers) _ ~ , ,, _ 13A Control 7.98 1.17 _ ~_ .
. 13B Nitrogen 8.l8 1.3l _ atmosphere _ . _ .
.

Example l l--Treatment of Polvstvrene Composite Membranes Composite polystyrene membranes, having intermediate layers of 15 poiytrimethylsilylpropyne and silicone on a microporous polysulfone support was prepared by the following procedure.
Ten parts of General Electric silicone RTV 615 Part A and one part Çeneral Electric silicone RTV 615 Part B were combined, pre-gelled by heating at 35C for 80 minutes, then diluted with heptane to form a 12 percent solution. The silicone solution was coated onto a 20 microporous polysulfone on polyester substrate membrane (obtained from FilmTec Corporation) using a gravure coating machine by the direct gravure method with a 180 Quadrangular roll, a line speed of 4 feeVmin. ( 1.2 meters/min.), an impression pressure of 10 psi. (69 kPa), and an oven temperature of 225F (107C).
A 0.75 percent by weight solution of polytrimethylsilylpropyne (PTMSP) in 25 heptane was prepared and filtered through glass wool . The solution was applied to the silicone/polysulfone suhstrate by di~coating on a gravure coating machine at a line speed of 2 feeVmin. (0.6 meters/min.) at an oven temperature of 225F ( 1 07C~.
A 2 percent by weight solution of polystyrene in amyl acetate was prepared and filtered through a 0.2~ polytetrafluoroethylene filter. The polystyrene solution was di~coated 30 ontothe PTMSP/silicone/polysulfone substrate using a small hand-operated coating machine.
The solvent was allowed to evaporate at room temperature and the membrane was then heated in a 60C oven overnight.
The membrane was UV irradiated for 1 or 3 minutes in a manner similar to that described in Exarnple 1. Data are reported in Table Xl.

^14-. . .

~ W094/L~6~ PCT~S93/1127S
21~9767 - `

TABLE XI
Polystyrene Composite Membranes . ..Oxyge- Flux Oxygen/
c~3 (stp) Nitrogen UV Time cm~sec cmHg Selectivity ~_ . - . . . - .
Before After Before After _ , _. , ,, . ___ - ... .
3 3.7 x 10-~ 3.6 x 10-~ 5.5 6.5 .~.. _. . _ .. ~ ,. . .
o 3 3 . 9 x 10 3 . 3 x 1 0--o S S S 9 1 g.4 x 10- 9.5 x lo- 5.4 6.0 , ,.. _. ~ f~' , , . .
7 1 3.6 x 10-" 3.7 X lo-o 5.7 6 9_ 1s Exam~le 12--Treatment of Tetrabromofluorene ContaininqPolycarbonate ComDosite Membranes -A composite membrane having a discriminating Jayer of tetrabromofluorene containing polycarbonate ~TBF PC), an intermediate layer of polytrimethyl-silylpropyne, and a support layer of microporous nylon was prepared.
The TBF PC may be prepared by the foliowing procedure. A three necked 0.5 liter round bottom flask equipped with a condenser, phosgene/nitrogen inlet, and a paddle stirrer connected to a Cole Parmer servodyne is charged with about 237 cubic centimeters of methylene thloride, 30.80 grams (0.046 moles) of 9,9-bis(3,5-dibromo-4-hydroxy-phenyl)fluorene, and 10.9 grams (0.138 moles) of pyridine. The resultant cle3r solution is stirred under 3 nitrogen atmosphere and about 4.6 grams (0.046 moles) of phosgene are bubbled into the reaction mixture over a peric~ of about 7 minutes. An additional quantity of about 1.0 gram (0.010 moles) of phosgene is bubbled in over about 18 minutes and the reaction mixture is stirred for about 16 hours. The reaction mixture is then scavenged with methanol, diluted with about 50 cubic centimeters of methylene chloride, washed twice with dilute hydrochloric àcid, arid then passed through DOWEX MSC-1 ion exchange resin. The polymer is isolated by adding the methylene chloride solution of polymer to a mixture of hexane/acetone. The precipitated polymer is dried under vacuum at about 1 20C for about 48 hours. The resultant polycarbonate is found to have an inherent viscosity of about 0.48 dUg at 25C in methylene chloride~
A 0.75 percent by weight solution of PTMSP in heptane was prepared and filtered through glass wool. The solution was applied to a 0.04 11 pore size Zetapor reinforced nylon substrate (obtained from CUNO Inc.) by dip-coating on a gravure coating machine at 2 WOg4/L~69 PCT~S93/11275 ~
~149767 feeVmin. (0.6 meterslmin.) line speed and 225F (107C) oven temperature. An 8 percent by weight solution of T8F PC in cyclopentanone was prepared and filtered through a 0.2~
polytetrafluoroethylene filter. The solution was hand cast onto the PTMSP/nylon substrate using a 1 mil doctor blade. The solvent was allowed to evaporate at room temperature and the 5 membrane was then heated in a 60~C oven overnight. The membrane was UV irradiated for 3 minutes in a manner similar to that described in Example 1. Data are reported in Table Xll.
TABLE XII
Tetrabromofluorene Polycarbonate Membranes _ -- . _ . .
Oxygen Flux Oxygen/
. c~3 (stp) Nitrogen UV Tlme cm sec cmHg Selectivity . _j , . . , . __ _ Before After Before After 3 4.~ x 10~ 5.0 x lO~ 5.9 11.3 . . . _ . - ~
3 ~.6 x 10 b 4.9 x lO~~ 6 1 . _ 3 6.3 x 10~ 6.5 x lO~ 5.9 7.5 .. .. . , . .- _ ~ --. _ . _ 3s

Claims

Claims:

1. A process for treating a gas separation membrane to improve its selectivity, characterized by the following steps:
A. first heating a membrane comprising a polymer having a UV excitable site and a labile protonic site in the polymeric backbone, such that a covalent bond is formed between said sites, at a temperature between 60 and 300°C
for a time sufficient to relax excess free volume in the polymer; and B. then irradiating the membrane with a UV radiation source in the presence of oxygen for a time sufficient to surface oxidize the membrane to obtain a treated membrane;
wherein the treated membrane exhibits at least a 10 percent increase in selectivity with less than a 60 percent decrease in permeability compared to the untreated membrane.

2. The process of Claim 1 wherein the membrane comprises an unsubstituted or substituted polycarbonate, polyestercarbonate, polyester, polystyrene, polysulfone, polyethersulfone, polyether, polyarylester,polyethylene terephthalate, cellulose ester, polybenzazole, polyurethane, or copolymer or physical blend thereof.

3. The process of Claim 2 wherein the membrane comprises an unsubstituted or substituted polycarbonate, polybenzazole, polyestercarbonate, polyester, polysulfone, polyethersulfone, polyether, polyarylester, or copolymer or physical blend thereof.

4. The process of Claim 3 wherein the membrane comprises polybenzazole or a polymer containing 9,9-bis(3,5-dibromo-4-hydroxyphenyl)fluorene, 9,9-bis(3,5-dichloro-4-hydroxyphenyl)fluorene, hexafluorobisphenol A, tetrahalohexafluorobisphenol A, or tetraalkyl-hexafluorobisphenol A moieties selected from the group consisting of polyether, polysulfone, polyethersulfone, polyarylester, polyester, polyestercarbonate, polycarbonate, and copolymers and physical blends thereof.

5. The process of Claim 2 wherein the membrane is a film.

6. The process of Claim 2 wherein the membrane is a hollow fiber.

7. The process of Claim 2 wherein the membrane is a composite.

8. The process of Claim 2 wherein the heating occurs at a temperature of between 100 and 250°C.

9. The process of Claim 8 wherein the UV irradiating occurs at a wavelength of between 180 and 400 nonmeters.

10. A gas separation membrane formed by the process of Claim 1.

11. A gas separation membrane comprising a thin discriminating layer characterized by a polymer selected from the group consisting of an unsubstituted or substituted polycarbonate, polyestercarbonate, polyester, polystyrene, polysulfone, polyethersulfone, polyether, polyarylester, polyamide, polyethylene teraphthalate, polybenzazole, polyurethane, or copolymer or physical blend thereof, which exhibits at least a 10 percent increase in selectivity with less than a 60 percent decrease in permeability after exposure to heat and UV irradiation to form a treated membrane compared to an untreated membrane.